Open Internet by MindsNet

Addressing Imbalanced Datasets in Epigenomics Analysis

The inherent biology of epigenomics data leads to imbalanced datasets, causing difficulties for AI models in predicting open and closed chromatin regions. This imbalance poses a significant challenge in applying deep learning to epigenomic sequence analysis. The issue persists across various stages of data analysis, including preprocessing, augmentation, and sampling.

Science, Biology, Molecular Biology